The Vice President of Engineering and Research at Google, Yossi Matias, discusses the potential of tech in reducing carbon emissions, flood forecasting, and more with Temitayo Jaiyeola, in this interview
How can we use technology to address environmental concerns?
The load of carbon emissions in cities is by cars in traffic lights, waiting for the light to turn green. It is worse even when it turns green, they accelerate and stop. And this go-and-stop kind of thing is contributing a lot to carbon emissions in cities.
Now, it turns out that we can do better, and the beauty is that without even installing any sensors or any infrastructure, just by analysing the data of traffic, we can come up with better scheduling of traffic lights. We provide this recommendation to the cities that can implement it in a very simple way by just adding a couple of seconds to one of the traffic lights, changing the amount of time that cars are at the intersection, or the amount of number of accelerations and stops. We have done this in places like Jakarta, Bangalore, Hamburg, and a few places in Africa. We have discovered this causes a 10 per cent decrease in carbon emission, which is quite significant obviously.
Also, floods are obviously one of the most devastating natural disasters, killing thousands of people per year, billions of economic damages impacting many hundreds of millions. So, we investigated how to use AI to predict floods. Surprisingly, we did a pretty good job, and it is now a solution. We announced a few months ago that we are now giving predictions, of up to seven days in advance, to over 80 countries worldwide, including 23 in Africa. We have made that available through flood hub.
The nature of this solution is based on building what is called a global hydrogen model. This is essentially taking historical data and then forecasting data of precipitation, building machine learning models that for any particular moment, any particular location, and with some data that we can rely on, we can provide forecast about what is going to be happening in the next few days and then we publish it in what in the flood hub, which is available for everyone to use.
Also, we provide this information via search and maps of Google. As it relates to Africa, we have a team in Accra and Nairobi, and they are working on various efforts around the climate crisis, and other things affecting the continent.
What has been the local uptake of your solution?
Everything that we do, we do in a very collaborative way with organisations and governments. Everything that we do is with the total knowledge of the local government and working closely with the World Meteorological Organisation. We are part of a programme that is called Early Warning For All, which is a global objective to provide early warning by 2027.
How can organisations in Nigeria, and Africa help communities prepare for floods?
People are more likely to act when organisations help spread the news about floods. This is not only in Africa. Red Cross is certainly a partner we work with and other NGOs as well. I think the government can help with organising this.
The way to think about how we can work with the community at large is in the collection of whatever can be helpful to build a model because public data is helpful in building this. Also, for the dissemination of information, we are currently using our platforms. We use flood hub, search, and maps. We have notifications. We are working on other means that we use for dissemination too.
However, organisations that can work with volunteers or leaders in various places can certainly be very helpful in disseminating information because, at the end of the day, people need to make decisions – what to do and when to do it. They need to apply the judgment of what it means that the water is going to be rising. Should they vacate their homes, or should they do something with their livestock or property? These are questions that need to be taken in a local way and having that community approach and support obviously or leadership by the government is obviously going to be very important.